Modeling Undergraduate Retention to Aid in Making Programming Decisions at Universities

• Activity Number: 256
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Education
• Author(s): Brianna Hitt*
• Keywords: logistic regression ; modeling ; student retention ; decision-making

Abstract

The primary role of the university is education. Through intentional programming, universities can meet the needs of all types of students: first generation, non-traditional, and low-income students, as well as first-time freshmen, athletes, and other groups. In order to improve decision-making as to the types of services offered to students before and after enrollment, we considered retention of undergraduate students in the College of Agricultural Sciences and Natural Resources at the University of Nebraska-Lincoln. We worked to model the retention of students from new student orientation to their first fall semester and beyond using logistic regression based on over 150 variables. Variables included all demographic and high school information, enrollment flags, and academic information for each semester. Additionally, we looked at correlations between variables and investigated shared characteristics of students by demographics and by the semester in which students were lost to attrition. The final model had a 97 percent success rate in predicting retention of individual students and will be used to aid in maintaining and developing programs to help ensure student success.